Archaeology - no digging required

16 June 2020

Interview with 

Martin Millett, University of Cambridge


Ancient Rome


The typical image of an archaeologist is somebody with a trowel down a hole, digging away, unearthing the past. Well it turns out that’s a bit old hat these days, because all you need is a quad bike with a fancy electronic gizmo on the back. The device is called a ground penetrating radar and, using one, scientists have mapped out the ruins of a Roman town in Italy, without ever having to pick up a trowel once! The town is Falerii Novi. It’s about 50 kilometres north of Rome and was founded over two thousand, two hundred years ago. All that’s on the site these days is the old town wall and a few sheep and it’s a protected monument, so you can’t dig there. Adam Murphy has been speaking to archaeologist Martin Millett, and he’s been explaining how the radar on the quad bike produced clear images of the entire town beneath the ground...

Martin - We've known about the site for a long time. It has had work done on it in the past; since the 18th Century, actually. We chose it for our work because of two things. One: it's greenfield site; in other words, the Roman town has gone and it's just fields now, so it's easy to access for experimental work. It's an historically important city anyway. The other point that makes it important from the point of view of this study is that, about 25 years ago, we did a magnetic survey of the whole site. By having that dataset, as well as the one that we collected now, we've got two geophysical datasets that could be directly compared. And that's very helpful for understanding how the techniques work and how they're complimentary.

Adam - Can you tell us how you went about looking for the things there that are under the ground?

Martin - Well it's relatively straightforward: we're using ground penetrating radar. So you have a radar antenna that sends a pulse of radio signal into the ground. And as the pulse of energy passes through changes in density in the ground - it hits a wall or something - you get an echo back; and the echo time is proportional to the depth. So at a basic level, you could just drag a single antenna across the site. What we're doing here, and what's new about this, is doing it over a very large area and at very high intensity. The solution for that is, rather than having a single radar antenna, you have a whole array of them; and our Belgian colleagues have put together 16 antennae that are dragged behind a quad bike. What that means is that you know the setup of the array, and with satellite surveying, you can tell exactly where each sensor is at any one time; and that means that you can collect a mass of radar data by gently dragging the array across the field.

Adam - How much data do you end up with at the other end of something like this?

Martin - Well, from this site it's somewhere around 28 billion data points.

Adam - And then how do you go about beginning to analyse 28 billion data points?

Martin - Well again, at the simple level, each single pulse gives you a series of reflections back. And it's a question of crunching the data to pull out the things that are coming back at the same nanosecond in time, so that you can differentiate what's going on at different depths beneath the surface. And then effectively you use an image processing software to turn that into a series of grayscale images that give you what's going on at individual depths; and there are various ways you can then play with that in three dimensional visualisation.

Adam - What kind of things have you found under the ground here?

Martin - We get the whole city plan. That's what we've been after. And that's of course made up of different buildings and different streets and structures. The beauty of this is that if you look at the images, they're very crisp, because we're collecting data at six and a quarter centimetre intervals; so you can see very small features, 20 to 30 centimetres across, under two metres of soil.

Adam - And you said you could see the individual buildings; what kind of buildings were they, there, that you've seen?

Martin - There... an array of big houses, and the more spectacular buildings. We've got very good evidence on the theatre, that was previously known, but we can see its structure very clearly now. We've got a completely hitherto unknown temple, and a big bath building, and a market building; and this very curious big monument up near the North gate, which is a large courtyard structure with colonnades on three sides, opened onto the street, but we don't know what it is! There's quite a lot of that sort of detective work to do on the images we've got at the moment.


Add a comment